A two-channel noise estimator for speech enhancement in a highly nonstationary environment

Min Seok Choi, Hong Goo Kang

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

This paper proposes a two-channel noise estimator for speech enhancement in a highly nonstationary environment. The proposed noise estimator utilizes a spatial filter which has a capability of extracting noise information even in a speech presence region. We exploit a first-order recursion method with timefrequency varying smoothing coefficients to accurately estimate a noise power spectral density (PSD) in both slowly and rapidly varying regions. The smoothing coefficients are determined by measuring the nonstationarity factor of noise, e.g., degree of noise variation. The nonstationarity factor is derived through a statistical assumption of stationary background noise, which does not need any assumption on the type of nonstationary noise. Since the proposed method efficiently estimates the noise PSD both in stationary and nonstationary regions, the enhanced speech obtained by applying the proposed algorithm to the two-channel enhancement system shows superior performance to conventional approaches in various noise environments.

Original languageEnglish
Article number5549863
Pages (from-to)905-915
Number of pages11
JournalIEEE Transactions on Audio, Speech and Language Processing
Volume19
Issue number4
DOIs
Publication statusPublished - 2011 Apr 6

Fingerprint

channel noise
Speech enhancement
Power spectral density
estimators
augmentation
smoothing
background noise
coefficients
estimates
filters

All Science Journal Classification (ASJC) codes

  • Acoustics and Ultrasonics
  • Electrical and Electronic Engineering

Cite this

@article{952fcd8bdf9c4d458ad8a56566f3c276,
title = "A two-channel noise estimator for speech enhancement in a highly nonstationary environment",
abstract = "This paper proposes a two-channel noise estimator for speech enhancement in a highly nonstationary environment. The proposed noise estimator utilizes a spatial filter which has a capability of extracting noise information even in a speech presence region. We exploit a first-order recursion method with timefrequency varying smoothing coefficients to accurately estimate a noise power spectral density (PSD) in both slowly and rapidly varying regions. The smoothing coefficients are determined by measuring the nonstationarity factor of noise, e.g., degree of noise variation. The nonstationarity factor is derived through a statistical assumption of stationary background noise, which does not need any assumption on the type of nonstationary noise. Since the proposed method efficiently estimates the noise PSD both in stationary and nonstationary regions, the enhanced speech obtained by applying the proposed algorithm to the two-channel enhancement system shows superior performance to conventional approaches in various noise environments.",
author = "Choi, {Min Seok} and Kang, {Hong Goo}",
year = "2011",
month = "4",
day = "6",
doi = "10.1109/TASL.2010.2066969",
language = "English",
volume = "19",
pages = "905--915",
journal = "IEEE Transactions on Speech and Audio Processing",
issn = "1558-7916",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "4",

}

A two-channel noise estimator for speech enhancement in a highly nonstationary environment. / Choi, Min Seok; Kang, Hong Goo.

In: IEEE Transactions on Audio, Speech and Language Processing, Vol. 19, No. 4, 5549863, 06.04.2011, p. 905-915.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A two-channel noise estimator for speech enhancement in a highly nonstationary environment

AU - Choi, Min Seok

AU - Kang, Hong Goo

PY - 2011/4/6

Y1 - 2011/4/6

N2 - This paper proposes a two-channel noise estimator for speech enhancement in a highly nonstationary environment. The proposed noise estimator utilizes a spatial filter which has a capability of extracting noise information even in a speech presence region. We exploit a first-order recursion method with timefrequency varying smoothing coefficients to accurately estimate a noise power spectral density (PSD) in both slowly and rapidly varying regions. The smoothing coefficients are determined by measuring the nonstationarity factor of noise, e.g., degree of noise variation. The nonstationarity factor is derived through a statistical assumption of stationary background noise, which does not need any assumption on the type of nonstationary noise. Since the proposed method efficiently estimates the noise PSD both in stationary and nonstationary regions, the enhanced speech obtained by applying the proposed algorithm to the two-channel enhancement system shows superior performance to conventional approaches in various noise environments.

AB - This paper proposes a two-channel noise estimator for speech enhancement in a highly nonstationary environment. The proposed noise estimator utilizes a spatial filter which has a capability of extracting noise information even in a speech presence region. We exploit a first-order recursion method with timefrequency varying smoothing coefficients to accurately estimate a noise power spectral density (PSD) in both slowly and rapidly varying regions. The smoothing coefficients are determined by measuring the nonstationarity factor of noise, e.g., degree of noise variation. The nonstationarity factor is derived through a statistical assumption of stationary background noise, which does not need any assumption on the type of nonstationary noise. Since the proposed method efficiently estimates the noise PSD both in stationary and nonstationary regions, the enhanced speech obtained by applying the proposed algorithm to the two-channel enhancement system shows superior performance to conventional approaches in various noise environments.

UR - http://www.scopus.com/inward/record.url?scp=79953287914&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79953287914&partnerID=8YFLogxK

U2 - 10.1109/TASL.2010.2066969

DO - 10.1109/TASL.2010.2066969

M3 - Article

AN - SCOPUS:79953287914

VL - 19

SP - 905

EP - 915

JO - IEEE Transactions on Speech and Audio Processing

JF - IEEE Transactions on Speech and Audio Processing

SN - 1558-7916

IS - 4

M1 - 5549863

ER -